This week’s roundup of data news highlights from May 23, 2026, to May 29, 2026, featuring scientists using satellite-linked sensors to detect animal behavior linked to poaching threats and a Boston Dynamics’ Atlas robot learning soccer-style movements by studying professional match footage.
1. Protecting Athletes’ Health
Professional hockey player Gabriel Landeskog has partnered with AI-driven movement platform Plantiga to monitor his movement mechanics as he returns from a long knee-injury layoff. The system’s in-shoe sensors capture hundreds of data points per second, tracking stride patterns, gait symmetry, and workload. Plantiga’s AI system analyzes those signals to flag subtle changes before they become injuries, helping Landeskog train safely and stay on the ice.
2. Tracking Animal Behavior
Scientists have used a satellite-based wildlife-tracking system called ICARUS to monitor animal movements at a scale that reveals how wildlife reacts to human threats. By pairing GPS tags with motion sensors, researchers can detect animals’ stress and panic responses. These behavioral signals train algorithms that alert rangers to potential poaching activity, using species such as zebras and giraffes as early-warning indicators to help protect vulnerable animals, such as rhinos.
3. Robotizing Farming
Oregon-based agrotech startup Canopii has built a compact automated greenhouse system designed to grow 40,000 pounds of greens annually using minimal land, water, and labor. Its robotic conveyors and seed-to-harvest automation handle planting, transplanting, and harvesting produce with almost no staff. The company aims to franchise these units so local operators can run efficient, low-cost farms that supply fresh, organic produce directly to nearby communities.
4. Accelerating Carbon Capture
University of Pennsylvania student Anya Draves has conducted a study exploring how AI could accelerate the development of carbon-capture technologies. Her research shows that AI can speed the design of carbon-absorbing materials and model efficient ways to store or reuse captured emissions. She argues that despite AI’s large energy demands, its ability to optimize carbon-capture systems could help reduce emissions and support long-term climate goals.
5. Automating Home Routines
Google Home has launched a new Gemini-powered automation feature that uses AI-analyzed camera activity to trigger smart-home routines. The system interprets what cameras detect, such as animals near trash bins, arriving mail, or a specific car entering the driveway, and activates preset actions automatically. The new feature also improves multi-action voice commands and speeds up Gemini’s responses for smoother smart-home control.
6. Learning Soccer
Boston Dynamics’ Atlas humanoid robot has learned soccer-style movements by watching match footage and practicing what it sees. The robot studies player positioning, reactions, and body mechanics on a large screen, then uses reinforcement learning and control systems trained in virtual environments to reproduce those motions. This combination of visual imitation and adaptive motor learning allows Atlas to guide a ball, run drills, and mimic celebrations.
7. Connecting to the Electrical Grid
U.S.-based tech company MeanderX has partnered with community solar developers to help them navigate Illinois’ heavily congested grid-connection process. By using AI to analyze utility data, the platform shows where the electrical grid has enough capacity for new solar and battery-storage projects. This visibility helps developers avoid bottlenecks, identify viable locations faster, and speed up decisions for new community-energy systems.
8. Navigating Rough Terrain
Researchers at Duke University have built a 20-legged robot called Argus that can move in any direction without relying on a fixed front or upright position. Using depth-sensing cameras, the robot navigates rough terrain, recovers from collisions, and continues operating even if some legs fail. Its design is based on a principle measuring how evenly robots accelerate in all directions, called dynamic isotropy, which could improve future search-and-rescue missions.
9. Decoding Medieval Text
Researchers at the University of Sweden have used machine learning to decode a medieval manuscript, revealing hidden messages about politics, diplomacy, and daily life. By training AI models on thousands of handwritten samples, the system can recognize obscure scripts, reconstruct damaged passages, and identify patterns in how scribes hid sensitive information. The work is helping historians gain new insight into medieval-society correspondences.
10. Enhancing Fans’ Experience
IBM has partnered with Formula 1 team Scuderia Ferrari to overhaul the team’s fan-engagement technology using an AI system and real-time race data. The collaboration focuses on rebuilding Ferrari’s fan app, transforming race telemetry, timing data, and fan activity into personalized stories, predictions, and interactive features. By analyzing how supporters interact with content, IBM’s AI system helps Ferrari create a more immersive experience for its growing fanbase.
